Landaise goose respiratory rate measuring method and system

Abstract

The invention relates to the technical field of millimeter wave radar detection, and discloses a Landaise goose respiratory frequency measurement method and system, and the method comprises the steps: obtaining an echo signal of a Landaise goose through a millimeter wave radar; performing time-frequency domain transformation processing on the echo signal to obtain a frequency spectrum feature of the target; performing target detection on the spectrum features, and extracting target micro-motion features of the Landaise geese; performing signal decomposition and reconstruction on the micro-motion features, and extracting the respiratory frequency of the Landaise geese; according to the method, the millimeter wave radar is used for assisting in judging the maturity of the goose fat liver, the respiratory rate of the Landaise goose can be monitored in real time, the individual health condition is judged, the vital sign state of the Landaise goose can be monitored while the Landaise goose is not disturbed, meanwhile, the labor cost can be saved, and the high-quality goose liver can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of millimeter-wave radar detection, and more specifically, it relates to a method and system for measuring the respiratory rate of Landes geese. Background Art

[0002] Landes geese are a famous special breed for foie gras in the world, originating from the Landes department in southwestern France. Their foie gras has a delicate texture and unique taste. In recent years, the goose industry in China has developed rapidly. Especially in Lu'an City, the breeding quantity of Landes geese has reached 4.041 million, and the development of the goose industry has shown a good momentum.

[0003] Currently, the judgment of foie gras maturity is the key bottleneck for obtaining high-quality foie gras. In the later stage of foie gras maturity, Landes geese will show physiological characteristics such as rapid breathing, red eyes, and listless mental state. If the foie gras cannot be obtained in time during this period, it will lead to the death of the geese and the scrapping of the foie gras; if the judgment is too early, it will result in insufficient maturity of the foie gras, small volume, and failure to meet the high-quality standard.

[0004] In the prior art, the state of Landes geese is mainly judged by workers through manual observation. This method has the following technical problems:

[0005] 1. It requires a large amount of human effort and energy, and the workload is huge;

[0006] 2. Workers need to continuously observe the states of a large number of Landes geese, and the average sleep time per person per day is less than two hours;

[0007] 3. Long-term fatigued work is prone to misjudgment and missed judgment;

[0008] 4. The manual observation method lacks quantitative basis, and the judgment result is highly subjective. Summary of the Invention

[0009] The present invention provides a method and system for measuring the respiratory rate of Landes geese, which solves the technical problems in the related art that continuously observing the states of a large number of geese requires a large amount of human effort and energy, and long-term fatigued work will lead to false alarms and missed alarms.

[0010] The present invention provides a method for measuring the respiratory rate of Landes geese, including the following steps:

[0011] Obtaining a signal: Using a millimeter-wave radar to obtain the echo signal of a Landes goose;

[0012] Two-dimensional Fourier transform: Performing a two-dimensional Fourier transform on the obtained received signal to obtain the spectral characteristics of the target;

[0013] Landes goose target detection: Performing target detection on the obtained spectral characteristics to extract the micro-motion characteristics of the Landes goose target;

[0014] Extract the breathing frequency of Landes geese: Decompose and reconstruct the micro-motion features to extract the breathing frequency of Landes geese.

[0015] Further, the two-dimensional Fourier transform includes the following steps:

[0016] Obtain the discrete difference frequency signals of all receiving channels within a frequency modulation period through ADC sampling;

[0017] Perform range dimension FFT operation on the discrete difference frequency signals of each receiving channel to obtain the spectral distribution of the target at different ranges;

[0018] Complete the range dimension FFT operation for all frequency modulation periods of one frame;

[0019] Arrange the one-dimensional FFT results in chronological order according to different transmitting channels;

[0020] Perform Doppler dimension FFT transform on the arranged data to obtain the Doppler-range dimension FFT spectrum, and determine the specific position of the target through the target range calculation formula. The formula for calculating the target range is:

[0021]

[0022] where, f b : difference frequency signal frequency;

[0023] T: modulation period;

[0024] B: frequency modulation bandwidth;

[0025] c: speed of light.

[0026] Further, the Landes goose target detection includes the following steps:

[0027] Count the number of cells occupied by the target main peak, and set range dimension and Doppler dimension protection cells;

[0028] Perform amplitude sorting on the reference cells, and take the average after removing the maximum and minimum values as the threshold;

[0029] Perform target detection according to the threshold, and extract the micro-motion features of the target. The micro-motion features include the range change, speed change, and acceleration change of the target;

[0030] Extract the spectrum of the target position in time series to form a time-spectrum diagram. Among them, the target detection adopts the constant false alarm rate detection algorithm. The constant false alarm rate detection algorithm maintains a constant false alarm probability by dynamically adjusting the detection threshold. The formula for calculating the detection threshold T is:

[0031]

[0032] where, α: threshold factor;

[0033] N: Number of reference units;

[0034] x i : Echo signal amplitude within the reference unit.

[0035] Furthermore, the steps for extracting the respiration rate of the Landes goose include the following:

[0036] Perform wavelet denoising on the original signal to remove some high-frequency noise;

[0037] Apply empirical mode decomposition to decompose the signal into multiple intrinsic mode functions, first perform IMF extraction, and then perform residual processing;

[0038] Select the components related to the respiration rate for reconstruction to obtain the respiration signal;

[0039] Perform peak detection on the reconstructed signal and calculate the respiration rate;

[0040] Among them, for wavelet denoising, select the Daubechies wavelet basis function for wavelet transform, set the decomposition level to 9, and decompose and reconstruct the signal containing the respiration rate range of the Landes goose.

[0041] Furthermore, the method for reconstructing the respiration signal is to first align the selected components in time and then perform linear superposition:

[0042]

[0043] Among them, s(t): the reconstructed respiration signal;

[0044] c i (t): the i-th component;

[0045] k: starting serial number;

[0046] n: ending serial number;

[0047] When calculating the respiration rate, first perform peak detection on the reconstructed signal, and calculate the reciprocal of the interval between adjacent peaks to obtain the instantaneous respiration rate. The formula for the instantaneous respiration rate is:

[0048]

[0049] Among them, f breath : instantaneous respiration rate;

[0050] T: time interval between adjacent peaks.

[0051] Furthermore, the specific steps for IMF extraction in the empirical mode decomposition include:

[0052] Identify the local extreme points of the radar received signal;

[0053] Connect the extreme points with cubic spline interpolation to form an envelope curve;

[0054] Calculate the average value m1(t) of the envelope curve to obtain the mean curve;

[0055] Screen the original signal and the average value of the envelope curve to check whether the IMF conditions are satisfied;

[0056] Among them, the formula for calculating the average value m1(t) of the envelope curve is:

[0057]

[0058] Among them, N: the number of sampling points of the radar received signal;

[0059] x(t i ): the i-th sampling point of the radar received signal;

[0060] When screening the original signal and the average value of the envelope curve, h1(t) is calculated by subtracting the local mean from the original signal, and the calculation formula is:

[0061] h1(t) = x(t) - m1(t);

[0062] Among them, x(t): the radar received signal;

[0063] h1(t): the screened signal;

[0064] Check whether h1(t) satisfies the IMF conditions, where the IMF conditions are that the number of extreme points of the signal h1(t) is less than 2 and the mean value of the signal h1(t) is 0.

[0065] Furthermore, the formula for calculating the residual in the residual processing of the empirical mode decomposition is:

[0066] r1(t)) = x(t) - c1(t);

[0067] Among them, c1(t): the IMF component, and its calculation formula is:

[0068]

[0069] Among them, N: the number of sampling points of the radar received signal;

[0070] h1(t i ): the i-th sampling point of the radar received signal;

[0071] Repeat the IMF extraction steps until the residual signal satisfies the termination condition, where the termination condition is that the number of extreme points of r1(t) is less than 2.

[0072] A Landes goose breathing frequency measurement system, comprising:

[0073] Domain control platform: It includes a signal source, a control platform and a digital processing platform, and is used to regulate the radar working mode and data processing;

[0074] Display and control platform: It is used to display the radar control interface and realize the visual control of the radar;

[0075] Millimeter-wave radar: Adopts a 3-transmit 4-receive system;

[0076] High-definition camera: It is used to obtain high-definition images, observe the state of Landes geese, and perform fusion processing with radar signals;

[0077] 5G wireless data terminal: It is used to form a wireless local area network and realize long-distance wireless regulation.

[0078] Furthermore, the millimeter-wave radar adopts the LFMCW system, uses the sawtooth wave modulation method, and includes a frequency modulation bandwidth, a carrier frequency start frequency, an intermediate frequency signal frequency, a rest period and a modulation period. The working parameters of the millimeter-wave radar include Doppler channels, range cells and range cell sizes. A positive frequency indicates that the target is moving away from the radar, and a negative frequency indicates that the target is approaching the radar. Specifically, it includes the following components:

[0079] Antenna system: It is used to transmit and receive millimeter-wave signals;

[0080] Transmitter: It converts an electrical signal into a millimeter-wave signal and transmits it. Among them, the transmitter includes a signal source, a power amplifier and an antenna array. Among them, the signal source generates a modulation signal, the power amplifier amplifies the signal power, and the antenna array transmits the millimeter-wave signal;

[0081] Receiver: It receives the millimeter-wave signal reflected by the target and converts it into an electrical signal. Among them, the receiver includes an antenna array, a low-noise amplifier and a signal processing unit. Among them, the antenna array receives the millimeter-wave signal, the low-noise amplifier amplifies the signal, and the signal processing unit processes and analyzes the received signal;

[0082] Signal processing unit: It processes and analyzes the received signal;

[0083] Data processing unit: It processes signal data and performs target detection and recognition;

[0084] Control unit: It is responsible for the overall control of the system, and the control parameters include antenna direction and transmission power.

[0085] A computer-readable storage medium is used to store computer-readable instructions, which can run the Landes goose breathing frequency measurement method as described above when read by a computer.

[0086] The beneficial effects of the present invention are as follows: By applying millimeter-wave radar to assist in the determination of the maturity of goose foie gras, it is possible to monitor the respiratory rate of Landes geese in real time without disturbing them, judge the individual health status, monitor the vital signs of Landes geese, screen out high-quality goose livers, and at the same time save labor costs and obtain high-quality goose livers. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 is a flowchart of the method for measuring the respiratory rate of Landes geese according to the present invention;

[0088] Figure 2 is the Doppler-range dimension FFT spectrogram according to the present invention;

[0089] Figure 3 is a schematic diagram of the constant false alarm detection design protection unit according to the present invention;

[0090] Figure 4 is the distance-spectrum diagram of the detection result of the Landes goose target according to the present invention;

[0091] Figure 5 is the time-spectrum diagram of the detection result of the Landes goose target according to the present invention;

[0092] Figure 6 is the principle block diagram of the radar system according to the present invention;

[0093] Figure 7 is a schematic diagram of the respiratory waveform curve of a Landes goose fed for 7 days during the experimental process according to the present invention;

[0094] Figure 8 is a schematic diagram of the respiratory rate curve of a Landes goose fed for 7 days during the experimental process according to the present invention;

[0095] Figure 9 is a schematic diagram of the respiratory waveform curve of a Landes goose fed for 17 days during the experimental process according to the present invention;

[0096] Figure 10 is a schematic diagram of the respiratory rate curve of a Landes goose fed for 17 days during the experimental process according to the present invention;

[0097] Figure 11 is a schematic diagram of the respiratory waveform curve of a Landes goose fed for 24 days during the experimental process according to the present invention;

[0098] Figure 12 is a schematic diagram of the respiratory rate curve of a Landes goose fed for 24 days during the experimental process according to the present invention;

[0099] Figure 13 is a scatter diagram of the daily respiratory rate of 11 Landes geese during the 27-day feeding process according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0100] Reference will now be made to exemplary embodiments to discuss the subject matter described herein. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein, and that changes may be made to the functions and arrangements of the elements discussed without departing from the scope of protection of the content of this specification. Each example may omit, substitute, or add various processes or components as needed. Additionally, the features described in some examples may be combined in other examples.

[0101] In at least one embodiment of the present invention, a method for measuring the respiratory rate of Landes geese is disclosed. As Figures 1 - 5 shown, it includes the following steps:

[0102] Step 1: Use a millimeter-wave radar to obtain the echo signal of the Landes goose. The millimeter-wave radar adopts a linear frequency modulation continuous wave system, and sends a linear frequency modulation continuous wave signal and receives the echo signal of the Landes goose;

[0103] Step 2: Perform a two-dimensional Fourier transform on the received signal to obtain the spectral characteristics of the target, specifically including:

[0104] Step 2-1: Obtain the discrete difference frequency signals of all receiving channels within a frequency modulation period T through ADC sampling;

[0105] Step 2-2: Perform a distance dimension FFT operation on the discrete difference frequency signal of each receiving channel to obtain the spectral distribution of the target at different distances.

[0106] Sampling frequency:

[0107] f s ≥2f max ;

[0108] Among them, f s is the sampling frequency, and f max is the maximum frequency of the signal;

[0109] Number of sampling points: N = 512;

[0110] Calculation of the target distance R:

[0111]

[0112] Among them, f b is the difference frequency signal frequency, T is the modulation period, B is the frequency modulation bandwidth, and c is the speed of light;

[0113] Step 2-3: Repeat steps 2-1 and 2-2 to complete the distance dimension FFT operation for all frequency modulation periods of one frame;

[0114] Step 2-4: Arrange the one-dimensional FFT results in chronological order according to the different transmitting channels;

[0115] Step 2-5: Perform Doppler-dimensional FFT transformation on the arranged data to obtain the Doppler-range dimensional FFT spectrum, specifically as follows Figure 2 shown. The parameters of the Doppler-range dimensional FFT spectrum are: 128 Doppler channels, 512 range cells, each range cell being 5.5 cm. Positive frequencies indicate that the target is moving away from the radar, and negative frequencies indicate that the target is approaching the radar.

[0116] Step 3: Perform target detection on the spectral features, extract the micro-motion features of the target of the Landes goose, and use an improved CFAR (constant false alarm rate detection) algorithm for target detection, including the following steps:

[0117] Step 3-1: Count the number of cells occupied by the target main peak and set the number of guard cells in the range dimension and Doppler dimension;

[0118] Step 3-2: Sort the amplitudes of the reference cells, remove the maximum and minimum values, and take the average as the threshold value;

[0119] Step 3-3: Perform constant false alarm rate detection. The calculation formula for the detection threshold T is:

[0120]

[0121] where α is the threshold factor, N is the number of reference cells, and X i is the echo signal amplitude within the reference cells.

[0122] Step 3-4: Extract the micro-motion features of the target of the Landes goose to distinguish it from other stationary targets. The micro-motion features of the target of the Landes goose include: the distance change of the target, the speed change of the target, and the acceleration change of the target.

[0123] Step 3-5: Extract the spectrum of the target position according to the time series to form a time-spectrum diagram;

[0124] Step 4: Perform signal decomposition and reconstruction on the micro-motion features, extract the breathing signal of the Landes goose, and use a method combining wavelet transform and EMD (empirical mode decomposition) to extract the breathing signal, including the following steps:

[0125] Step 4-1: Analyze the breathing signal and waveform features of the through-wall radar, select the appropriate wavelet basis function Daubechies, set the decomposition level to 9, and decompose and reconstruct the signal containing the breathing frequency range of the Landes goose;

[0126] Step 4-2: Apply empirical mode decomposition (EMD) to decompose the radar received signal into a finite number of intrinsic mode functions (IMFs). First, perform IMF extraction, and then perform residual processing, specifically including:

[0127] Step 4-2-1: The specific steps of IMF extraction include:

[0128] Identify the local extreme points of the radar received signal;

[0129] Connect the extreme points with cubic spline interpolation to form an envelope;

[0130] Calculate the average value m1(t) of the envelope to obtain the mean curve;

[0131] Screen the original signal and the average value of the envelope to check whether the IMF conditions are satisfied;

[0132] The formula for calculating the average value m1(t) of the envelope is:

[0133]

[0134] where N is the number of sampling points of the radar received signal, and x(t i ) is the i-th sampling point of the radar received signal;

[0135] When screening the original signal and the average value of the envelope, h1(t) is calculated by subtracting the local mean from the original signal, and the calculation formula is:

[0136] h1(t) = x(t) - m1(t);

[0137] where x(t) is the radar received signal and h1(t) is the screened signal

[0138] Check whether h1(t) satisfies the IMF conditions, where the IMF conditions are that the number of extreme points of the signal h1(t) is less than 2 and the mean value of the signal h1(t) is 0;

[0139] Step 4-2-2, residual processing: First calculate the residual r1(t), and then repeat Step 4-2-1 until the residual signal satisfies the termination condition. The termination condition is that the number of extreme points of r1(t) is less than 2. The formula for calculating the residual is:

[0140] r1(t) = x(t) - c1(t);

[0141] where c1(t): IMF component, and its calculation formula is:

[0142]

[0143] where N is the number of sampling points of the radar received signal, and h1(t i ) is the i-th sampling point of the radar received signal;

[0144] Repeat the IMF extraction steps until the residual signal satisfies the termination condition, where the termination condition is that the number of extreme points of r1(t) is less than 2.

[0145] Step 4-3: Reconstruct the respiratory signal from the time domain using the IMF components that reflect the structural characteristics of the vital signs. First, align the selected IMF components in time, and then perform linear superposition. Let t be the time variable, and the calculation formula for the reconstructed respiratory signal s(t) is as follows:

[0146]

[0147] where s(t) is the reconstructed respiratory signal, c i (t) is the i-th IMF component, k is the starting IMF serial number (usually select the lowest-frequency IMF containing the respiratory frequency), and n is the ending IMF serial number (usually select the highest-frequency IMF containing the respiratory frequency).

[0148] Step 4-4: Apply EMD to extract the respiratory signal. The important structures of the respiratory signal are concentrated in the low-frequency part. Only use a part of the IMF components that reflect the spectral structural characteristics of the respiratory signal to reconstruct the respiratory signal. Perform peak detection on the reconstructed signal s(t), calculate the time interval between adjacent peaks, take the reciprocal of the time interval to obtain the instantaneous respiratory frequency, and take the average of the instantaneous frequencies within a certain time window to obtain the final respiratory frequency measurement result. Among them, the calculation formula for the instantaneous respiratory frequency is:

[0149]

[0150] where f breath is the instantaneous respiratory frequency, and T is the time interval between adjacent peaks.

[0151] Here, the present invention provides an implementation example:

[0152] Select a standard breeding room with an area of 20 square meters as the test environment. Keep the environmental temperature at 23±2°C and the relative humidity at 60±5%. Conduct long-term monitoring on 11 Landes geese for 27 days. Measure their respiratory frequencies every day during the 27-day feeding process. The scatter plot is as Figure 13 shown. Due to excessive fright, the respiratory rates of some individual Landes geese are rapid. Exclude such errors.

[0153] Select 3 healthy Landes geese from the above samples as test objects, which are respectively at the 7th day, 17th day, and 24th day of feeding. The specific detection results are as Figures 7 - 12 shown.

[0154] Landes geese fed for 7 days: The respiratory frequency is about 0.16 Hz, and they breathe 10 times per minute;

[0155] Landes geese fed for 17 days: The respiratory frequency is about 0.32 Hz, and they breathe 19 times per minute;

[0156] Force-fed Landes geese for 24 days: Respiratory rate is about 0.38 Hz, breathing 23 times per minute.

[0157] The experiment concludes that: as the force-feeding days of geese increase, the liver of Landes geese becomes larger, the breathing becomes more rapid, and the respiratory rate rises; it is possible to monitor the respiratory rate of Landes geese in real time without disturbing them, judge the individual health status, monitor the vital signs of Landes geese, screen out high-quality goose livers, and at the same time save labor costs and obtain high-quality goose livers.

[0158] As Figure 6 shown, in an embodiment of the present invention, a respiratory rate measurement system for Landes geese includes:

[0159] Domain control platform: It includes a signal source, a control platform, and a digital processing platform, and is used to regulate the radar working mode and data processing;

[0160] Display control platform: It is used to display the radar control interface and realize the visual control of the radar;

[0161] Millimeter-wave radar: Adopts a 3-transmit 4-receive system;

[0162] High-definition camera: It is used to obtain high-definition images, observe the state of Landes geese, and perform fusion processing with radar signals;

[0163] 5G wireless data terminal: It is used to form a wireless local area network and realize long-distance wireless regulation.

[0164] Furthermore, the millimeter-wave radar adopts the LFMCW system and the sawtooth wave modulation method, and includes a frequency modulation bandwidth, a carrier frequency start frequency, an intermediate frequency signal frequency, a rest period, and a modulation period. The working parameters of the millimeter-wave radar include Doppler channels, range cells, and range cell sizes. A positive frequency indicates that the target is moving away from the radar, and a negative frequency indicates that the target is approaching the radar. Specifically, it includes the following components:

[0165] Antenna system: It is used to transmit and receive millimeter-wave signals;

[0166] Transmitter: It converts an electrical signal into a millimeter-wave signal and transmits it. Among them, the transmitter includes a signal source, a power amplifier, and an antenna array. Among them, the signal source generates a modulation signal, the power amplifier amplifies the signal power, and the antenna array transmits the millimeter-wave signal;

[0167] Receiver: It receives the millimeter-wave signal reflected by the target and converts it into an electrical signal. Among them, the receiver includes an antenna array, a low-noise amplifier, and a signal processing unit. Among them, the antenna array receives the millimeter-wave signal, the low-noise amplifier amplifies the signal, and the signal processing unit processes and analyzes the received signal;

[0168] Signal processing unit: processes and analyzes the received signals; specifically, the signal processing unit includes range - dimension FFT, Doppler - dimension FFT, target detection, and micro - motion feature extraction. Among them, the range - dimension FFT converts the received signals into range - dimension FFT spectra, the Doppler - dimension FFT converts the range - dimension FFT spectra into Doppler - dimension FFT spectra, the target detection is used to detect the target position and velocity, and the micro - motion feature extraction is used to extract the micro - motion features of the target;

[0169] Data processing unit: processes the signal data and performs target detection and recognition; specifically, the data processing unit includes target detection and micro - motion feature extraction;

[0170] Control unit: is responsible for the overall control of the system, and the control parameters include antenna direction and transmission power.

[0171] In an embodiment of the present invention, a computer - readable storage medium is used to store computer - readable instructions, which can run the Landes goose breathing frequency measurement method as described above when read by a computer.

[0172] The above describes the embodiments of the present invention. However, the embodiments are not limited to the above - mentioned specific implementation manners. The above - mentioned specific implementation manners are merely illustrative rather than restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make more equivalent embodiments in various forms, all of which fall within the protection scope of this embodiment.

Claims

1. A method for measuring the respiratory rate of Landes geese, characterized in that: The following steps are involved: Acquiring signals: Using millimeter-wave radar to acquire the echo signal of the Landes goose; Two-dimensional Fourier transform: Perform two-dimensional Fourier transform on the received signal to obtain the spectrum characteristics of the target; Landes goose target detection: Target detection is performed on the obtained spectrum features to extract the target micro-motion features of the Landes goose; Extracting the breathing frequency of the Landais goose: Perform signal decomposition and reconstruction on the micro-motion features to extract the breathing frequency of the Landais goose.

2. A method for measuring the respiratory rate of Landes geese according to claim 1, characterized in that: The two-dimensional Fourier transform comprises the following steps: The discrete difference frequency signals of all receiving channels within a frequency modulation period are obtained through ADC sampling; Perform distance dimension FFT operation on the discrete difference frequency signal of each receiving channel to obtain the spectrum distribution of the target at different distances; Complete the distance dimension FFT operation of all frequency modulation cycles in one frame; Arrange the one-dimensional FFT results in chronological order according to the different transmission channels; Perform Doppler-dimensional FFT transformation on the arranged data to obtain the Doppler-range-dimensional FFT spectrum. The specific position of the target is determined by the target distance calculation formula. The target distance calculation formula is: Among them, f b : difference signal frequency; T: modulation period; B: FM bandwidth; c: speed of light.

3. A method for measuring the respiratory rate of Landes geese according to claim 1, characterized in that: The Landes goose target detection comprises the following steps: Count the number of cells occupied by the target main peak and set the range dimension and Doppler dimension protection units; The reference units are sorted by amplitude, and the maximum and minimum values ​​are removed and the average is taken as the threshold value; Detect the target based on the threshold value and extract the target's micro-motion features, which include the target's distance change, speed change, and acceleration change; The frequency spectrum of the target position is extracted in time series to form a time-spectrum diagram, wherein the target detection adopts a constant false alarm detection algorithm, which maintains a constant false alarm probability by dynamically adjusting the detection threshold. The calculation formula of the detection threshold T is: Where, α: threshold factor; N: number of reference units; x i : Echo signal amplitude within the reference unit.

4. A method for measuring the respiratory rate of Landes geese according to claim 1, characterized in that: The extraction of the respiratory rate of the Landes goose comprises the following steps: Perform wavelet denoising on the original signal to remove some high-frequency noise; Apply empirical mode decomposition to decompose the signal into multiple intrinsic mode functions, first perform IMF extraction, and then perform residual processing; Select the components related to the respiratory frequency for reconstruction to obtain the respiratory signal; Perform peak detection on the reconstructed signal and calculate the respiratory rate; Among them, wavelet denoising selects Daubechies wavelet basis function for wavelet transform, sets the decomposition layer number to 9, and decomposes and reconstructs the signal containing the breathing frequency range of Landes goose.

5. A method for measuring the respiratory rate of Landes geese according to claim 4, characterized in that: The method for reconstructing the respiratory signal is to first align the filtered components in time and then perform linear superposition: s(t): reconstructed respiratory signal; c i (t): the i-th component; k: starting sequence number; n: end sequence number; When calculating the respiratory frequency, the reconstructed signal is firstly peak detected, and the inverse of the interval between adjacent peaks is calculated to obtain the instantaneous respiratory frequency, wherein the calculation formula of the instantaneous respiratory frequency is: Among them, f breath : instantaneous respiratory rate; T: time interval between adjacent peaks.

6. A method for measuring the respiratory rate of Landes geese according to claim 4, characterized in that: The specific steps of IMF extraction in the empirical mode decomposition include: Identify the local extreme points of the radar received signal; Use cubic spline interpolation to connect the extreme points to form an envelope; Calculate the envelope average value m1(t) to obtain the mean curve; Screen the original signal and the envelope average to check whether they meet the IMF conditions; The formula for calculating the envelope average value m1(t) is: Where, N: number of sampling points of radar receiving signal; x(t i ): the i-th sampling point of the radar receiving signal; When the original signal and the envelope mean are screened, h1(t) is calculated by subtracting the local mean from the original signal, using the formula: h1(t)=x(t)-m1(t); Where, x(t): radar received signal; h1(t): filtered signal Check whether h1(t) satisfies the IMF condition, where the IMF condition is that the number of extreme points of the signal h1(t) is less than 2 and the mean of the signal h1(t) is 0.

7. A method for measuring the respiratory rate of Landes geese according to claim 6, characterized in that: The residual calculation formula of the residual processing in the empirical mode decomposition is: r1(t)=x(t)-c1(t); Where, c1(t): IMF component, which is calculated as: Where, N: number of sampling points of radar receiving signal; h1(t i ): the i-th sampling point of the radar receiving signal; The IMF extraction step is repeated until the residual signal meets the termination condition, where the termination condition is that the number of extreme points of r1(t) is less than 2.

8. A system for measuring the respiratory rate of a Landai goose, used to implement a method for measuring the respiratory rate of a Landai goose according to any one of claims 1 to 7, characterized in that: include: Domain control platform: includes signal source, control platform and digital processing platform, used to control radar working mode and data processing; Display and control platform: used to display the radar control interface and realize visual control of the radar; Millimeter wave radar: adopts 3-transmit 4-receive system; High-definition camera: used to obtain high-definition images, observe the status of the Landes goose, and fuse them with radar signals; 5G wireless data terminal: used to form a wireless local area network and realize long-distance wireless control.

9. A Landes goose respiratory rate measurement system according to claim 8, characterized in that: The millimeter wave radar adopts the LFMCW system and the sawtooth wave modulation method, including the frequency modulation bandwidth, the carrier frequency start frequency, the intermediate frequency signal frequency, the rest period and the modulation period. The working parameters of the millimeter wave radar include the Doppler channel, the distance unit and the distance unit size. The positive frequency indicates that the target is far away from the radar, and the negative frequency indicates that the target is close to the radar. Specifically, it includes the following components: Antenna system: used to send and receive millimeter wave signals; Transmitter: converts the electrical signal into a millimeter wave signal and sends it. The transmitter includes a signal source, a power amplifier and an antenna array. The signal source generates a modulated signal, the power amplifier amplifies the signal power, and the antenna array sends the millimeter wave signal. Receiver: receives the millimeter wave signal reflected by the target and converts it into an electrical signal, wherein the receiver includes an antenna array, a low-noise amplifier and a signal processing unit, wherein the antenna array receives the millimeter wave signal, the low-noise amplifier amplifies the signal, and the signal processing unit processes and analyzes the received signal; Signal processing unit: processes and analyzes the received signal; Data processing unit: processes signal data and performs target detection and recognition; Control unit: responsible for the overall control of the system, and the control parameters include antenna direction and transmission power.

10. A computer-readable storage medium, characterized in that: It is used to store computer-readable instructions, and when the computer-readable instructions are read by a computer, the method for measuring the respiratory rate of a Landes goose as described in any one of claims 1 to 7 can be executed.

Images